A three-dimensional (3D) microscope for patterned substrate measurement can include an objective lens, a reflected illuminator, a transmitted illuminator, a focusing adjustment device, an optical sensor, and a processor. The focusing adjustment device can automatically adjust the objective lens focus at a plurality of Z steps. The optical sensor can be capable of acquiring images at each of these Z steps. The processor can control the reflected illuminator, the transmitted illuminator, the focusing adjustment device, and the optical sensor. The processor can be configured to capture first and second images at multiple Z steps, the first image with the pattern using the reflected illuminator and the second image without the pattern using one of the reflected illuminator and the transmitted illuminator.
|
1. A method of measuring a patterned substrate sample, the patterned substrate sample including a plurality of patterned substrate features, the method comprising:
varying a relative distance between the patterned substrate sample and an objective lens at predetermined steps;
at one or more of the predetermined steps:
projecting an image of a patterned article onto a focal plane of the objective lens;
capturing a first image with a pattern associated with the patterned article and the patterned substrate sample, and storing the first image in a first image array; and
capturing a second image of the patterned substrate sample without the pattern associated with the patterned article, and storing the second image in a second image array;
using the second image array to roughly estimate a bottom position of the patterned substrate features;
using the roughly estimated bottom position and the second image array to identify first locations of the patterned substrate sample that do not include the plurality of patterned substrate features;
using the first image array and the first locations to accurately determine the bottom position of the patterned surface features;
using the accurately determined bottom position and the second image array to identify second locations of the patterned substrate sample that include the plurality of patterned substrate features;
determining a top of each patterned substrate feature based on the second locations and one of the first image array and the second image array; and
calculating geometric parameters of patterned substrate features using the second locations, the accurately determined bottom position, and the top of each patterned substrate feature.
2. The method of
3. The method of
4. The method of
5. The method of
6. The method of
7. The method of
8. The method of
9. The method of
10. The method of
11. The method of
12. The method of
13. The method of
14. The method of
15. The method of
|
This application is a continuation of U.S. patent application Ser. No. 13/172,686, entitled “#D Microscope And Methods Of Measuring Patterned Substrates”, filed Jun. 29, 2011 which claims priority to Provisional Application 61/367,352, entitled “3D Microscope And Methods Of Measuring Patterned Substrates”, filed Jul. 23, 2010, and incorporated by reference herein. This application is also related to U.S. Pat. No. 7,729,049, entitled “3D Optical Microscope”, which issued Jun. 1, 2010, U.S. Pat. No. 7,944,609, entitled “3D Optical Microscope”, which issued on May 17, 2011, U.S. Pat. No. 8,174,762 entitled “3D Optical Microscope” which issued May 8, 2012, and U.S. Pat. No. 8,184,364, entitled “Illuminator For A 3D Optical Microscope”, which issued May 22, 2012, all of which are incorporated by reference herein.
Field of the Invention
The present invention relates to an optical microscope and in particular to a three-dimensional (3D) microscope and methods of measuring a patterned substrate (PS) in 3D.
Description of the Related Art
High Brightness Light Emitting Diode (HBLED) has generated tremendous interest among research communities and various industries due to its reliability, long lifetime, and environmental benefits when compared to conventional light sources. Typically, conventional HBLEDs are manufactured on transparent substrates such as sapphire, silicon carbide, and other materials. To improve light extraction efficiency, manufacturers often roughen the substrate surface to form patterns so that a greater portion of light generated in the active layer can be emitted.
U.S. Pat. No. 6,657,236, entitled “Enhanced Light Extraction In LEDs Through The Use Of Internal And External Optical Element”, which issued to Thibeault on Dec. 2, 2003, and U.S. Pat. No. 7,384,809, entitled “Method Of Forming Three-Dimensional Features On Light Emitting Diodes For Improved Light Extraction”, which issued to Donofrio on Jun. 10, 2008, disclose methods of creating various repeating patterns on a silicon carbide substrate to enhance the light extraction efficiency of a HBLED. As described in U.S. Pat. No. 7,384,809, images from a secondary electron microscope (SEM) can be used to verify the shapes of these patterned substrates.
U.S. Pat. No. 7,683,386, entitled “Semiconductor Light Emitting Device With Protrusions To Improve External Efficiency And Crystal Growth”, which issued to Tanaka on Mar. 23, 2010, U.S. Pat. No. 7,745,245, entitled “Semiconductor Light Emitting Device”, which issued to Niki on Jun. 29, 2010, and U.S. Published Application 2008/0067916, entitled “Light Emitting Device Having A Patterned Substrate And The Method Thereof”, which was filed by Hsu on Jul. 30, 2007, teach various ways to generate repeating patterns on a sapphire substrate. In these references, SEM images are provided to confirm the quality of the patterned sapphire substrates.
U.S. Pat. No. 7,704,763, entitled “Highly Efficient Group-III Nitride Based Light Emitting Diodes Via Fabrication Of Features On An N-Face Surface”, which issued to Fuji on Apr. 27, 2010, discloses a method of manufacturing a HBLED on a sapphire substrate, then using laser lift-off to de-bond the substrate from the diode structure. At this point, an etch process can be used to create random pyramids on an N-face GaP surface to achieve a roughened surface. Again, images from a SEM can be used in monitoring formation of the random pyramid features.
As part of the manufacturing process development and process control, manufacturers need to measure the geometry of the pattern on the substrates. These measurements typically include the shape, height, size, pitch, and space of the pattern features. Although a conventional SEM can image various patterned features, it cannot measure height information. As a result, cross-sectional SEM (x-SEM) has become the standard metrology tool in the HBLED industry. However, x-SEM is a destructive method, which requires breaking of a HBLED prior to taking a measurement. In addition, x-SEM measurement has to be carried out in a vacuum environment and therefore is slow in throughput. Furthermore, an x-SEM system is expensive to buy and maintain.
Non-destructive, non-contact optical systems have been used in the semiconductor industry for years in measuring masks on transparent substrates. For example, U.S. Pat. No. 6,323,953, entitled “Method And Device For Measuring Features On A Transparent Substrate”, which issued to Blaesing-Bangert on Nov. 27, 2001, and U.S. Pat. No. 6,539,331, entitled “Microscopic Feature Dimension Measurement System”, which issued to Fiekowsky on Mar. 25, 2003, teach methods for accurately measuring a line width on a photomask using an optical microscope setup. However, these methods can only measure line width, i.e. lateral dimensions, and cannot provide accurate height information.
Therefore, a need arises for a non-destructive method that is accurate, easy to use, and relatively inexpensive to measure and monitor patterned substrates. The need is met with the present invention which will be explained in the following detailed description.
A three-dimensional (3D) microscope for patterned substrate measurement can include an objective lens, a reflected illuminator, and a transmitted illuminator. The reflected illuminator can be configured to provide reflected light for a patterned substrate sample and to project an image of a patterned article onto and remove the image of the patterned article from a focal plane of the objective lens. The transmitted illuminator can be configured to provide transmitted illumination for the patterned substrate sample.
The 3D microscope can also include a focusing adjustment device, an optical sensor, and a processor. The focusing adjustment device can automatically adjust the objective lens focus at a plurality of Z steps. The optical sensor can be capable of acquiring images at each of these Z steps. The processor can control the reflected illuminator, the transmitted illuminator, the focusing adjustment device, and the optical sensor. The processor can be configured to capture first and second images at multiple Z steps, the first image with the pattern using the reflected illuminator and the second image without the pattern using one of the reflected illuminator and the transmitted illuminator.
In one embodiment, the patterned article is a piece of glass with a pre-determined pattern thereon. The optical sensor can include a charge-coupled device (CCD) camera or a complementary metal-oxide semiconductor (CMOS) camera. The transmitted illuminator can be a light emitting diode (LED) and one of a lens and a lens group. The focusing adjustment device can be a motorized mechanical Z stage or a piezo Z stage. The motorized Z stage can include a lead screw or a ball screw coupled to a linear bearing. The piezo Z stage can be mounted on a sample chuck or a microscope turret.
A method of designing a 3D microscope for measurement of a patterned substrate is also described. This method includes providing the above-described components.
A method of measuring a patterned substrate sample is also described. A patterned substrate sample is defined as including a plurality of patterned substrate features. In this method, a relative distance between the patterned substrate sample and an objective lens can be varied at predetermined steps. At one or more of the predetermined steps, the following additional steps can be performed.
An image of a patterned article can be projected onto a focal plane of the objective lens. A first image with a pattern associated with the patterned article and the sample can be captured and then stored in a first image array. The image of the patterned article can then be removed from the focal plane of the objective lens. A second image of the sample without the pattern associated with the patterned article can be captured and then stored in a second image array
A first mask can be generated to roughly distinguish the patterned substrate features from a background area of the patterned substrate sample. This first mask is based on the second image array. A second mask can be generated to accurately distinguish the patterned substrate features from the background area. This second mask is based on the first image array and the first mask.
A top of each patterned substrate feature can be determined using the second mask and one of the first image array and the second image array. Geometric parameters of patterned substrate features can be calculated using the second mask and the top of each patterned substrate feature.
Capturing the second image can include using a reflected illuminator or a transmitted illuminator. The transmitted illuminator can be a light emitting diode (LED) and one of a lens and a lens group. Generating the first mask can include using one of color, intensity, or a combination of both color and intensity. Generating the second mask can include using a thresholding method. The geometric parameters can include size, pitch, height, space, and top size of the patterned substrate features. Varying the relative distance between the patterned substrate sample and the objective lens can include using a motorized mechanical Z stage or a piezo Z stage. The motorized Z stage can include a lead screw or a ball screw coupled to a linear bearing. The piezo Z stage can be mounted on a sample chuck or a microscope turret.
In one embodiment, the method of measuring a patterned substrate sample can include automatically varying the relative distance between the patterned substrate sample and the objective lens. This automatically varying can include a first auto-focus technique and a second auto-focus technique. The first auto-focus technique can include a conditional early exit. This conditional early exit can include determining whether more than a threshold scan range is done. When more than the threshold scan range is done, then a standard deviation can be calculated from accumulated contrast values, otherwise scanning can continue. The conditional early exit can further include determining whether the maximum contrast value is more than a specified minimum early exit threshold and a current step contrast is less than a maximum contrast by at least the standard deviation. If so, then an early exit is approved, otherwise not. The first auto-focus technique can also include capturing images while the Z stage is moving between scan steps, thereby allowing a speed of the first auto-focus to be as fast as a camera frame rate.
In one embodiment, the second auto-focus technique can have a step size smaller than that of the first auto-focus technique. The second auto-focus technique can include detecting a falling contrast pattern. This falling contrast pattern can be a scan step with peak contrast, followed at least a plurality of scan steps of lower contrast values. In one embodiment, the plurality of scan steps is four scan steps.
The term “patterned substrate” as used herein describes a roughened surface. This roughened surface can be formed on any transparent substrate used in the HBLED industry, e.g. sapphire, silicon carbide, GaP, etc. Embodiments of patterned substrates can use repeating features or random features.
In one embodiment, the optical components of illuminator 100 can be mounted inside a dark enclosure with two openings (not shown), e.g. a top opening and a bottom opening. The top opening can be directly above beam-splitter 107 while the bottom opening can be directly below beam-splitter 107. These two openings allow light from both light paths to interact with the outside world.
As described in further detail below, after hitting beam splitter 107, the light from one of the first and second sources travels through an objective lens and then hits the sample surface. Reflected light travels back through the objective lens, beam splitter 107, and a coupling lens (not shown). A camera receives this reflected light and forms an image (see, e.g.
In a preferred embodiment, light sources 101 and 102 can include light emitting diodes (LEDs); however, other light sources such as halogen lamps, fiber-coupled lights, lasers, etc can also be used and are within the scope of this invention. Note that although lenses 105 and 106 are described as being an achromat doublet lens and a double-convex lens, those skilled in the art will understand that other types of lenses can also be used and are within the scope of this invention.
Note that patterned article 104 can be piece of glass, photographic film, or other transparent material that is capable of carrying the pattern. The patterned surface of patterned article 104 is located at the effective focal plane of the lens group including lenses 105 and 106. As described in further detail below, patterned article 104 can be used in illuminator 100 to project an image of the pattern onto the focal plane of an objective lens to create enough contrast so that 3D height information of a sample (e.g. the patterned substrate) can be obtained.
A microscope objective lens 210 is mounted on a turret 260. Turret 260 can hold at least one objective lens and is mounted directly below a bottom opening of illuminator 100. When light source 101 or 102 is turned on, the lens group including lenses 105 and 106 projects an image of the light source onto the entrance pupil of microscope objective lens 210, thereby ensuring uniform illumination of a sample 220. Moreover, when light source 102 is turned on, the lens group including lenses 105 and 106 project an image of the pattern on patterned article 104 onto the focal plane of objective lens 210.
Positioning means 230 (shown as a double-headed arrow for simplicity) is provided to change the relative position between sample 220 and objective lens 210. As a result, different features on sample 220 can be brought into focus of objective lens 210. In a preferred embodiment, positioning means 230 can include a motorized Z stage or piezo Z stage. In other embodiments, other ways to vary the relative position between sample 220 and objective lens 210 can be used. For example, objective lens 210 could be mounted on a piezoelectric actuator, thereby allowing sample 220 to remain stationary while objective lens 210 moves up and down. Positioning means 230 can also include a manual or motorized XY stage (not shown), thereby allowing sample 220 to be moved in a horizontal plane. Therefore, positioning means 230 can provide an XYZ range of motion. Those skilled in the art will recognize variations of the described positioning means 230.
Coupler 240 in conjunction with objective lens 210 yields an image of sample 220 on an optical sensor 250. In a preferred embodiment, optical sensor 250 can be either a charge-coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) camera. Coupler 240 could be of a single magnification or of a variable magnification depending on patterned substrate sample types. For example, coupler 240 could contain a 1× lens and a 2× lens mounted on a linear slider.
Light source 280 provides transmitted illumination for sample 220. As such, light source 280 is called a transmitted illuminator. In a preferred embodiment, light source 280 is an LED. In other embodiments, light sources such as halogen lamps, fiber coupled lights, lasers, and etc can be used. Sample 220 can sit on a chuck 270, which is formed from either a transparent glass plate or a metal plate with a through hole in the middle to allow light from light source 280 to go through. A processor 290 can be used to control positioning means 230, illuminator 100, light source 280, and an optical sensor 250. Processor 290 can also analyze data and create a 3D image of sample 220. In one embodiment, processor 290 can include a personal computer.
When the region on sample 220 to be measured is moved under objective 210, certain LEDs of the array in proximity to the measured region can be turned on to provide the transmitted illumination. Piezo stage 510 allows for precise vertical movement of sample 220. Note that piezo stage 510 provides one possible embodiment of positioning means 230 (
In another embodiment providing high Z movement resolution, a piezo Z drive can be mounted onto lead screw/ball screw 610. In this configuration, the illuminator, the objective turret, and the objective lenses can then be moved to a nominal focus position by lead screw/ball screw 610.
A 3D microscope system can employ two methods of camera control for data and image acquisition. In a first method, for every scan the system turns on one of the first and second light sources, moves the Z stage to the desired position, and issues a trigger signal to the camera to acquire the image. Once image data is transferred from the camera to the computer memory, the system switches to the other light source (if needed) and issues another trigger signal to the camera. The system then moves the Z stage to the next position and repeats the process until the number of Z steps is completed.
In a second method, for every scan, the system moves the sample from a starting position to an ending position in a continuous motion without stopping. The camera trigger signals are generated from either the position of the encoder counts of the motor used in the lead screw/ball screw mechanism, or from the position sensor of the Z-drive or Z-stage. The system electronics then send out the trigger signals at equal distant intervals to the camera to capture the image. The interval between each trigger is programmed to match with the transfer rate of the camera. The system continuously transfers the data to the PC memory until the stage completes its motion. Note that in the second method, the system turns on one of the first and second light sources at the start of a scan and does not switch light sources during the scan. If a second pass is required, then the system runs another continuous motion scan using the other light source.
Due to sample thickness variations, different locations on the patterned substrate sample may have different Z positions relative to the objective lens. In addition, a patterned substrate sample is not flat but has surface texture, i.e. a vertical profile. Therefore, before each patterned substrate measurement, the point to be measured on the sample needs to be focused. This focusing can be done manually, but the precision or repeatability of the resulting start point can vary. To minimize this variation of the start point for a repeatable patterned substrate measurement, an auto-focus procedure can be used to start scanning from a consistent starting Z position.
Note that a simple, conventional method to search for the best focus would be to command the Z stage to step through the whole search range at a predefined step size, and at each step, wait until the Z motion settles, command the camera to capture an image, and then wait for the image data to arrive. After the image contrast from all steps is analyzed, the Z position corresponding to the highest contrast could be determined. The position of highest contrast would be the best guess focus Z position. While this simple method works and is accurate, it is undesirably slow.
An auto-focus technique in accordance with the present invention can take advantage of the 3D microscope image contrast from the projected pattern as well as from the sample itself. When part of the sample surface is brought close to the focal plane, the corresponding part of the image contrast gets higher and will reach a peak when that part of the sample surface is at the focal plane. The auto-focus technique described herein has two-passes: the first pass being optimized for speed and the second pass being optimized for accuracy.
TABLE 1
Frame rate (fps)
Step size (microns)
Offset (microns)
1 to 15
up to 0.5
0
1 to 15
more than 0.5
1
16 to 60
up to 0.5
1
16 to 60
more than 0.5
2
faster than 60
up to 0.5
3
faster than 60
more than 0.5
5
This offset can be used to generate a more accurate first pass best guess focus Z position (described in further detail in reference to
In step 804, the Z stage is moved to a second pass starting Z position. In one embodiment, the second pass starting Z position is calculated to be at half of the second pass scan range below the first pass best guess focus Z position.
In step 805, parameters for the second pass auto-focus can be set. For example, in one embodiment, the step size can be set to be half of that in the first pass to improve resolution. Moreover, the second pass scan range can be set to N times the first pass step size, wherein N is a positive integer or fraction. In one embodiment, the scan range is set by a user. In another embodiment, the scan range is set by the recipe, which is specific to a particular sample and system configuration. An optimized choice for the second pass scan range may be determined by the accuracy of the first pass best guess focus Z position. The second pass will be slow if its scan range is set too large. However, if the scan range is set too small, the actual focus may not be covered, thereby potentially missing the true best focus. In one embodiment, the second pass scan range may be larger or smaller than 8 times the first pass step size.
In step 806, the second pass auto-focus can be executed to generate a second pass best guess focus Z position. In step 807, the Z stage can be moved to a best guess focus Z position plus the final Z offset. This positioning can ensure that the patterned sample measurement can start from a consistent specific point.
Notably, in step 903, Z stage movement can be triggered when image frames are captured rather than arrival at designated scan steps. Thus, images can be captured while the Z stage is moving between scan steps, thereby allowing the auto-focus speed to be as fast as the camera frame rate. To speed up each scan step, the first pass auto-focus can run the camera at its fastest frame rate, which occurs in a free running, continuous capture mode where images are continuously captured and transferred to processor 290 (
Because the algorithm does not wait for the Z stage to finish its motion, the actual Z position corresponding to the image received will be less than the commanded Z position. This difference is typically small, if the camera frame rate is slow, but becomes more significant when the camera is fast. Because of this difference between commanded and actual Z position, the best guess focus Z position corresponding to the commanded Z position will probably be shifted from the actual best guess focus Z position. Therefore, a calibration offset table linking this shift with the step size and camera frame rate (see, step 803,
Step 905 determines whether a focus has been found (described in further detail in reference to
Step 921 of this early exit technique can set a default flag of no early exit (i.e. early exit is FALSE). Step 922 can determine whether more than a threshold scan range (i.e. a minimum number of scan steps) is done. For example, in one embodiment, if less than half of the scan steps are scanned, then step 923 continues scanning and subsequently returns to step 922. If more than half of scan steps are scanned, then step 924 can calculate the standard deviation (sigma) from the accumulated contrast values.
Step 925 can determine whether the maximum contrast value is more than a specified minimum early exit threshold and the contrast value of current scan step is at least one standard deviation below that of the maximum contrast value of the accumulated contrast values. For a typical image with a contrast value between 0 and 1000, a threshold of 10 would be reliable for most samples. If so, then a focus is found and step 926 can set the early exit flag to TRUE. If not, then a focus is not found and step 927 can retain the early exit flag setting of FALSE. Step 928, which follows either step 926 and or step 927, can return to the first pass technique with an appropriate flag for step 905.
Because the focus position is expected to be within the second pass starting and ending positions, the saved contrast values in the second pass are expected to have a rise and fall pattern. Therefore, at step 934, a simple check of falling contrast values can be done to determine if focus is found, thereby indicating that the second pass auto-focus is done. Notably, because the second pass scan step size is smaller than that of the first pass, the rise and fall pattern may not be sharp (for example, several scan steps may have the same or similar maximum contrast value, thereby not changing contrast values significantly). Also, because the search range of the second pass only needs to cover the uncertainties of the first pass, it can be much smaller than that of the first pass. As a result, the number of search steps in the second pass steps is small. In the preferred embodiment, the maximum number of second pass search steps is 19. Statistics calculated on such limited number of contrast values may not be meaningful. Therefore, in one embodiment, a falling contrast value can be defined to be a scan step with peak contrast, followed by 4 scan steps of lower contrast values. A value of less than 4 makes the second pass stop sooner, and a value of more than 4 makes the second pass more accurate.
If a falling contrast pattern is detected, then step 935 can calculate the best guess focus Z position. In one embodiment, the best guess focus Z position can be the middle of the Z position range corresponding to the middle of the maximum contrast range.
As described above, a two-pass patterned sample measurement technique can include first and second passes. The first pass auto-focus can advantageously stop the focus search early without going through all the steps. Moreover, the images can be captured while the Z stage is moving between scan steps, thereby allowing the auto-focus speed to be as fast as the camera frame rate. To further improve upon the first pass autofocus accuracy, the second pass can search at a smaller step size within a small range around the best guess focus Z position found in the first pass.
Additional improvements on the two-pass auto-focus technique can also be provided. For example, in one embodiment, different criteria can be used for determining maximum contrast or best focus. That is, instead of calculating the overall contrast of the whole image, the contrast of a portion of the image, or the contrast of several different portions of the image can be calculated and then used to determine the maximum contrast for the best focus.
Because the goal of the auto-focus procedure is to position the patterned substrate sample at a consistent point to start the patterned sample measurement, other means, including different auto-focus methods, such as auto-focus method using image intensity with a confocal optical setup, or using focus signal from a separate focus sensor, instead of image contrast as described above, can be used in other embodiments to achieve the same results. Such variations are within the scope of this invention.
In yet another embodiment, a user can also specify the position found from the auto-focus algorithm to represent the middle, the bottom, or the top of scan range (or any position in between). This specificity is needed because the auto-focus algorithm will find the best focus consistently at the highest average contrast surface. The highest contrast surface position can be at the base, the middle, or the top of the patterned sample structure depending on its shape and composition. This extra control can be specified in the GUI/recipe, thereby tailoring the GUI/recipe for different patterned sample wafers or chips.
For certain patterned substrate features, e.g. such as cone and triangular features, non-patterned images are collected with the transmitted illumination (see, e.g. embodiments of
Step 1002 can calculate a histogram spread for non-patterned image array 1001 as an indication of either color or intensity distribution. Note that the spread of the histogram tends to be at the maximum when the Z position is around the bottom of the patterned substrate features, i.e. at the substrate. Therefore, the Z position of the substrate can be roughly determined by obtaining the maximum of the histogram spread. Step 1003 can generate a first mask by applying a threshold indicated by the median of the histogram to non-patterned image array 1001 at this Z position. Note that the first mask is a binary mask having an opaque background with transparent features that roughly represent the bases of the patterned substrate features. Note that this binary mask is implemented in software and is not a physical mask.
The first mask can be used to roughly distinguish the patterned substrate features from the substrate. The boundary between the patterned substrate features and the substrate may not be very accurate because of noise introduced in part from the actual shape of the patterned substrate features.
Step 1005 can use patterned image array 1004 and the first mask (from step 1003) to calculate the contrast on the substrate at this Z position. As noted above, the patterned substrate feature designations of the first mask typically have noise issues. Therefore, step 1005 can use the first mask to eliminate those areas from the contrast calculation. As a result, the contrast calculation with patterned image array 1004, but without the roughly designated patterned substrate features, allows an accurate Z position of the substrate to be determined. In one embodiment, this accurate Z position can be based on the maximum contrast value.
Step 1006 can generate an accurate binary mask, called a second mask, by applying a threshold indicated by the median of the histogram to non-patterned image array 1001 at this Z position. This second mask can be used to accurately distinguish the patterned substrate features from the substrate.
Step 1007 can use the second (i.e. accurate) mask and either the non-patterned image array 1001 or the patterned image array 1004 to calculate the top and the Z position of each of the patterned substrate features. Information from the recipe (described below in reference to
Step 1008 can calculate the size, pitch, height, space, and top size of each patterned substrate feature based on the top and Z position values computed in step 1007 and the second mask. Size can be defined as a diameter for a circular patterned substrate feature or a height for a patterned substrate having an equilateral triangular base. Pitch can be defined as a distance between the center of a current patterned substrate feature and the center of a neighboring patterned substrate features. In one embodiment, the average of the distances between all neighboring patterned substrate features can be calculated and used as the pitch. Height can be determined by the absolute difference of the maximum Z position and the minimum Z position within the patterned substrate feature. Space can be defined as the difference between the pitch and the size. Note that the top size only applies to a patterned substrate feature having a flat plane on its top. Statistics of these values, such as average, median, standard deviation and others can be obtained over the whole field of view to get more reliable readings of the patterned substrate features. Step 1009 can report these values.
Compared to prior art, the 3D microscope system described herein has several advantages. Specifically, the 3D microscope system is easy to use, is based on a non-contact, non-destructive method, offers a low cost of ownership among a class of existing patterned sample measurement tools, notably, provides fast, accurate, and repeatable measurement on key parameters that matter to patterned substrate manufacturers. Among these parameters are the size, pitch, height, and space of patterned substrate features.
The embodiments described herein are not intended to be exhaustive or to limit the invention to the precise forms disclosed. As such, many modifications and variations will be apparent. Accordingly, it is intended that the scope of the invention be defined by the following Claims and their equivalents.
Kudinar, Rusmin, Hou, Zhen, Xu, James Jianguo, Soetarman, Ronny, Lee, Ken Kinsun, Nguyen, Hung Phi, Stainton, James Nelson
Patent | Priority | Assignee | Title |
Patent | Priority | Assignee | Title |
2977847, | |||
3437395, | |||
4198571, | Apr 28 1977 | National Research Development Corporation | Scanning microscopes |
4629324, | Dec 29 1983 | Robotic Vision Systems, Inc. | Arrangement for measuring depth based on lens focusing |
4692051, | Dec 06 1982 | SCOTT, KEN W | Hot water heater knob adaptor |
4852985, | Oct 16 1986 | Olympus Optical Co., Ltd. | Illuminating device for microscopes |
4995716, | Mar 09 1989 | VISION OPTIMIZATION, LTD | Method and apparatus for obtaining the topography of an object |
5022743, | Mar 27 1987 | The Board of Trustees of the Leland Stanford Junior University | Scanning confocal optical microscope |
5151609, | Aug 02 1989 | Hitachi, Ltd. | Method of detecting solid shape of object with autofocusing and image detection at each focus level |
5184021, | Jun 24 1991 | ISHIKAWAJIMA-HARIMA HEAVY INDUSTRIES CO , LTD | Method and apparatus for measuring the dimensions of patterned features on a lithographic photomask |
5381236, | Feb 12 1991 | Oxford Sensor Technology Limited | Optical sensor for imaging an object |
5528033, | Mar 29 1995 | International Business Machines Corporation | Automatic surface profiling for submicron device |
5867610, | Feb 18 1992 | TRIPATH IMAGING, INC | Method for identifying objects using data processing techniques |
5932872, | Jul 07 1994 | Q3DM, LLC | Autofocus system for scanning microscopy having a volume image formation |
6088155, | Jun 04 1996 | Carl Zeiss Microscopy GmbH | Device for switching the operating modes of a microscope tube |
6275335, | Jul 16 1999 | SL3D, Inc.; SL3D, INC | Single-lens 3D method, microscope, and video adapter |
6323953, | Apr 30 1998 | Leica Microsystems Semiconductor GmbH | Method and device for measuring structures on a transparent substrate |
6376818, | Apr 03 1998 | OXFORD UNIVERSITY INNOVATION LIMITED | Microscopy imaging apparatus and method |
6539331, | Sep 08 2000 | KLA-Tencor Corporation; KLA-Tencor Technologies Corporation | Microscopic feature dimension measurement system |
6616291, | Dec 23 1999 | AQUA LIGHTING TECHNOLOGIES, LLC | Underwater lighting assembly |
6657236, | Dec 03 1999 | Cree, Inc | Enhanced light extraction in LEDs through the use of internal and external optical elements |
7384809, | Apr 01 2004 | CREE LED, INC | Method of forming three-dimensional features on light emitting diodes for improved light extraction |
7683386, | Aug 19 2003 | Nichia Corporation | Semiconductor light emitting device with protrusions to improve external efficiency and crystal growth |
7704763, | Dec 09 2003 | Japan Science and Technology Agency | Highly efficient group-III nitride based light emitting diodes via fabrication of structures on an N-face surface |
7729049, | May 26 2007 | KLA-Tencor Corporation | 3-d optical microscope |
7745245, | Jul 24 2001 | Nichia Corporation | Semiconductor light emitting device |
7944609, | May 26 2007 | KLA-Tencor Corporation | 3-D optical microscope |
8174762, | May 26 2007 | KLA-Tencor Corporation | 3-D optical microscope |
20030060809, | |||
20040210402, | |||
20040257360, | |||
20050046702, | |||
20060007533, | |||
20060038144, | |||
20070109633, | |||
20070171519, | |||
20070212049, | |||
20080067916, | |||
20080291532, | |||
20080291533, | |||
20090078888, | |||
20090237676, | |||
20100074489, | |||
20100134595, | |||
20100135573, | |||
20110110567, | |||
CN101034198, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Jun 28 2011 | HOU, ZHEN | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 28 2011 | XU, JAMES JIANGUO | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 28 2011 | LEE, KEN KINSUN | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 28 2011 | STAINTON, JAMES NELSON | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 28 2011 | NGUYEN, HUNG PHI | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 28 2011 | KUDINAR, RUSMIN | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
Jun 29 2011 | SOETARMAN, RONNY | ZETA INSTRUMENTS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 038554 | /0323 | |
May 09 2016 | KLA-Tencor Corporation | (assignment on the face of the patent) | / | |||
Aug 01 2018 | ZETA INSTRUMENTS, INC | KLA-Tencor Corporation | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 046530 | /0323 |
Date | Maintenance Fee Events |
Aug 14 2017 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
Aug 19 2022 | M1551: Payment of Maintenance Fee, 4th Year, Large Entity. |
Date | Maintenance Schedule |
Feb 19 2022 | 4 years fee payment window open |
Aug 19 2022 | 6 months grace period start (w surcharge) |
Feb 19 2023 | patent expiry (for year 4) |
Feb 19 2025 | 2 years to revive unintentionally abandoned end. (for year 4) |
Feb 19 2026 | 8 years fee payment window open |
Aug 19 2026 | 6 months grace period start (w surcharge) |
Feb 19 2027 | patent expiry (for year 8) |
Feb 19 2029 | 2 years to revive unintentionally abandoned end. (for year 8) |
Feb 19 2030 | 12 years fee payment window open |
Aug 19 2030 | 6 months grace period start (w surcharge) |
Feb 19 2031 | patent expiry (for year 12) |
Feb 19 2033 | 2 years to revive unintentionally abandoned end. (for year 12) |